The intern's noise survey report was concerning. 'The reception area measures -42 dBFS,' it read. 'The server room is at -18 dBFS.' The project manager stared at the numbers with the expression of someone who suspects they're being pranked but isn't sure enough to laugh.

The intern, fresh from audio engineering school, had measured the noise using a USB audio interface and a DAW. The meter in the DAW reported levels in dBFS — decibels relative to full scale, where 0 dBFS is the maximum digital level before clipping. This is a useful reference for recording. It is not a useful reference for acoustic noise measurement.

A reading of -42 dBFS told you exactly nothing about the actual sound pressure level in the room. It told you what the signal level was relative to the ADC's clipping point, which depended on the preamp gain, which the intern had set to 'whatever it was already on.' At one gain setting, -42 dBFS might correspond to 60 dB SPL. At another, 90 dB SPL. Without calibration, the number was meaningless.

The server room at -18 dBFS might have been 85 dBA or 65 dBA or anything in between. The intern had measured with precision — the numbers were reported to one decimal place — but without accuracy. Precise nonsense is still nonsense.

The Moral: dBFS and dB SPL are different reference systems. SonaVyx's SPL meter with proper calibration workflow converts digital levels to acoustic levels using a known reference point — because a number without the right unit and reference is just a number.